[gmx-users] NVT or NPT ensemble?

[Hendrik Preuss]

Hi all,

I need your help to decide which ensemle to use for my simulation.
I generated a system consisting of a GPCR in a POPC bilayer (81
monomers) that is solvated with spc water (in a triclinic box). 10 Cl
cations were included to have total charge of 0. I work with the ffgmx
FF and the lipid.itp and popc.itp from the group of Dr. Tieleman.

First I tried the following settings in my mdp file for 500 ps
equillibration (protein fixed):
;electrostatics and vdw
coulombtype              = PME
vdwtype                  = Cut-off
rcoulomb                 = 1.4
rvdw                     = 1.0
DispCorr                 = EnerPres
optimize_fft             = yes
;
;temperature coupling
tcoupl                   = berendsen
tc-grps                  = Protein      POP  SOL Cl
tau_t                    = 0.1  0.1  0.1 0.1
ref_t                    = 300  300  300 300
;
;pressure coupling
pcoupl                   = berendsen
tau_p                    = 1.0
pcoupltype               = isotropic
compressibility          = 4.5e-5
ref_p                    = 1.0

With these settings the volume of my box decreased from 404 nm^3 to 347
nm^3 during the first 40 ps (the pressure of the system at the beginning
was -2495 bar). During 500 ps the thickness of the lipid bilayer is
diminished such that the transmembrane domains are no longer surrounded
by lipid but water.

Therefore I decided to keep the volume constant in a new run (pcoupl =
no).
But then, the lipids changed to disordered conformations and regions
with "empty space" appeared. In this case the pressure fluctuates
between -500 and -1000 bar.

Do you have an idea what to do? What is the correct ensemble type in my
case?
Thanks a lot for your help!

Hendrik

*************************************************

Hendrik Preuss
Institute of Pharmacy
University of Regensburg
D-93040 Regensburg, Germany
Phone: +49 941 943 3329
e-mail: hendrik.preuss at chemie.uni-regensburg.de

************************************************